Air pollution effects on
myocardial infarction
Efeitos da poluição do ar no infarto
do miocárdio
ABSTRACT
OBJECTIVE: Myocardial infarction is an acute and severe cardiovascular disease
that generally leads to patient admissions to intensive care units and few cases are initially admitted to infirmaries. The objective of the study was to assess whether estimates of air pollution effects on myocardial infarction morbidity are modified by the source of health information.
METHODS: The study was carried out in hospitals of the Brazilian Health System
in the city of São Paulo, Southern Brazil. A time series study (1998-1999) was performed using two outcomes: infarction admissions to infirmaries and to intensive care units, both for people older than 64 years of age. Generalized linear models controlling for seasonality (long and short-term trends) and weather were used. The eight-day cumulative effects of air pollutants were assessed using third degree polynomial distributed lag models.
RESULTS: Almost 70% of daily hospital admissions due to myocardial infarction
were to infirmaries. Despite that, the effects of air pollutants on infarction were higher for intensive care units admissions. All pollutants were positively associated with the study outcomes but SO2 presented the strongest statistically significant association. An interquartile range increase on SO2 concentration was associated with increases of 13% (95% CI: 6-19) and 8% (95% CI: 2-13) of intensive care units and infirmary infarction admissions, respectively.
CONCLUSIONS: It may be assumed there is a misclassification of myocardial
infarction admissions to infirmaries leading to overestimation. Also, despite the absolute number of events, admissions to intensive care units data provides a more adequate estimate of the magnitude of air pollution effects on infarction admissions.
KEYW O RD S: Air pollution, adverse effects. Cardiovascular diseases. Time-series study.
RESU M O
OBJETIVO: O infarto do miocárdio é uma doença cardiovascular grave que tem
como indicação a internação em unidades de terapia intensiva, com poucos indicados para admissão em enfermarias. O objetivo do estudo foi investigar se as estimativas dos efeitos da poluição atmosférica nas internações por infarto do miocárdio são modificadas de acordo com a fonte de informações de saúde.
MÉTODOS: Em hospitais do Sistema Único de Saúde (SUS), na cidade de São
Paulo, foi realizado estudo de séries temporais (1998-1999) tendo como desfechos as internações por infarto em unidades de terapia intensiva e em enfermarias, em pessoas acima de 64 anos. Foram utilizados modelos lineares generalizados, controlados para sazonalidade (de longa e curta duração) e variáveis climáticas. Foram construídos
Sônia CendonI
Luiz A A PereiraI
Alfésio L F BragaI
Gleice M S ConceiçãoI
Abraão Cury JuniorI
H élio RomaldiniI
Antônio C LopesI
Paulo H N SaldivaII
I Departamento de Clínica Médica.
Faculdade de Medicina. Universidade Federal de São Paulo. São Paulo, SP, Brasil
I I Laboratório de Poluição Atmosférica.
Faculdade de Medicina. Universidade de São Paulo. São Paulo, SP, Brasil
Correspondence:
Sônia Cendon
Rua Afonso Braz, 656 2o andar
modelos distribuídos de defasagem polinomial de terceiro grau, para avaliar os efeitos acumulados nos oito dias anteriores à exposição.
RESULTADOS: Aproximadamente 70% das internações por infarto no miocárdio
ocorreram em enfermarias. Apesar disso, os efeitos da poluição sobre os casos foram maiores nas internações em unidades de terapia intensiva. Todos os poluentes mostraram uma associação positiva com os desfechos, mas o SO2 apresentou uma associação mais robusta e estatisticamente significante. O aumento do intervalo interquartil para as concentrações observadas do SO2 foi associado ao aumento em 13% (IC 95%: 6-19) e 8% (IC 95%: 2-13) nas internações em unidade de terapia intensiva e enfermarias, respectivamente.
CONCLUSÕES: Pode-se supor que exista um erro de classificação das internações
por infarto nas enfermarias, superestimando o número de internações. No entanto, o menor número de internações por infarto do miocárdio em unidades de terapia intensiva, é o indicador mais adequado para estimar os efeitos da poluição atmosférica nas internações por infarto.
D ESCRITO RES: Poluição do ar, efeitos adversos. D oenças cardiovasculares, estudo de séries temporais.
INTRODUCTION
The incidence and prevalence of cardiovascular dis-eases have increased in the last decades and became one of the main causes of death among adults.3,18 In
metropolitan areas, increasing levels of air pollut-ants constitute an additional risk factor for those dis-eases.22,24 Most of the epidemiological studies
quan-tifying air pollution effects on human health have focused on mortality events, reporting both chronic and acute strong and robust associations.2,5,18,21
The interest in investigating cardiovascular disease morbidity has increased in the last years. Lin et al13
recently studied the association between air pollution and ischemic cardiovascular emergency room visits in São Paulo, Brazil. Environmental CO was associated with angina and acute myocardial infarction in a short lag time. Santos et al20 conducted a study, also in São
Paulo, following 48 healthy traffic workers daily ex-posed to eight hours of urban air pollution, evaluating blood pressure and heart rate. The results showed in-creased blood pressure and dein-creased heart rate vari-ability during the highest polluted periods. These ef-fects are in agreement with other reports on cardiovas-cular effects of urban air pollution exposure.4,9,19
Despite the severity of cardiovascular events, most patients will not present fatal outcomes and will be cared at emergency room services and/or admitted to intensive care units or infirmaries. These morbid-ity outcomes are highly relevant in order to get a precise estimative of cardiovascular air pollution health effects.
In the present study, it was assessed the effects of daily variation of urban air pollutants on myocardial infarction (MI) hospital admissions using data from intensive care units (ICU) and regular infirmaries of the Brazilian Health System (SUS) hospitals.
M ETH O D S
Hospital admissions due to MI, classified according to the ICD-10,25 were obtained from the Public Health
Data Analysis System Division (DATASUS) for 112 SUS hospitals in the city of São Paulo, Southern Bra-zil, between January 1998 to December 1999. It was selected the daily number of admissions due to MI in two admission sectors: infirmaries and ICU.
The Companhia de Tecnologia de Saneamento
Ambiental (São Paulo State Environmental
(University of São Paulo Institute of Astronomy and Geophysics, IAG-USP).
In order to evaluate the association between air pol-lution and hospital admissions due to MI, general-ized additive Poisson regression distributed lag mod-els8,21 were applied. The two study outcomes were the
number of admissions due to MI in elderly admitted to two hospital units: ICU and infirmaries. The ex-planatory variables were smooth function of time, temperature and humidity and dummy variables for days of the week. Based in previous studies,13 it was
employed two-day moving average for temperature and humidity.
It was considered the LOESS smoother, available in S-Plus™ 2000 software, using more stringent con-vergence parameters in order to assure concon-vergence of its iterative estimation procedures.6 The
smooth-ing parameter for time was chosen in such a manner as to control the seasonality, removing long-term trends out of the data and minimizing the autocorrelation of the residuals. If there were any re-maining autocorrelation, autoregressive terms were added to the model.
The effect of air pollution on mortality is distributed over time, i.e., the number of admissions in a single day may be due to the exposure to air pollution not only on that day but also over the preceding days.
Considering that, it was used third degree polyno-mial distributed lag models (PDL) to investigate a 8-day cumulative effect (lag 0 through lag 7) because it can provide a biologically plausible lag structure.2
The sum of the effect of a unit increase in air pollu-tion on a single day is the sum of regression coeffi-cients for each day (β0 + …+ β7). The standard errors
of the estimates are given by the variance of the sum of β0 + …+ β7, adjusting for over dispersion. It was
estimated percentage increases and 95% confidence intervals (CI) in MI due to interquartile range increases in air pollutants concentrations.
RESU LTS
Table 1shows the descriptive analysis of the main variables included. Almost 70% of the daily hospital admissions due to MI were to infirmaries. None of air pollutants had daily concentrations above the Bra-zilian air quality standards.*
Table 2 presents the Pearson correlations coefficients of the main variables. Both infirmaries and ICU ad-missions had positive significant correlations with pollutants, except O3, and several pollutants were correlated to each other.
Table 3 shows the percentage increase and 95% CI for hospital admissions to ICU and infirmaries due to MI. SO2 presented a positive significant association
*Brasil, Conselho Nacional do Meio Ambiente. Resolução no 3, de 28 de junho de 1990. Diário Oficial da União, Brasília (DF), 22 ago 1990,
Seção I, p. 15937-9.
Table 1 - Descriptive analysis of the main variables. São Paulo, 1998-1999.
Variable D aily mean SD 5% IQ R 95% N
MI admissions
Intensive care unit 7.90 4.63 1.00 7.00 16.00 717
Infirmaries 18.50 12.51 3.00 16.00 45.00 724
W eather
Temperature (oC) 15.21 3.42 9.80 5.40 20.60 724
Humidity (%) 81.09 8.34 64.78 10.30 93.00 724
Pollutants
SO2 14.75 8.16 5.31 10.00 31.57 724
NO2 100.19 42.88 48.01 54.67 189.77 724
O3 72.10 38.69 24.78 50.23 145.13 724
PM10 48.34 21.41 25.73 22.50 95.89 724
CO 2.93 1.39 1.30 1.42 5.92 724
MI: Myocardial infarction SD: Standard deviation; IQR: Interquartile range; N: Days of observations
Table 2 - Pearson correlation of the main variables. São Paulo, 1998-1999.
Vari abl e Infi rmary ICU Temperature H umidity SO2 NO2 O3 PM10
Infirmary 1.00
ICU 0.64** 1.00
Temperature -0.03 -0.08* 1.00
H umidity 0.02 -0.08* 0.02 1.00
SO2 0.28** 0.35** -0.34** -0.44** 1.00
NO2 0.30** 0.31** -0.05 -0.34** 0.70** 1.00
O3 -0.09* 0.03 0.05 -0.37** 0.31** 0.40** 1.00
PM10 0.21** 0.23** -0.23** -0.54** 0.77** 0.76** 0.41** 1.00
CO 0.18** 0.16** -0.04 -0.36** 0.52** 0.65** 0.00 0.69**
with both hospital admissions dependent variables. The percentage increase of ICU admissions was 58% greater than infirmaries admissions. An interquartile range increase on SO2 concentration was associated with increases of 13% (95% CI: 6-19) and 8% (95% CI: 2-13) on ICU and infirmary MI admissions, re-spectively. ICU admissions were also associated with O3 (9.3% increase). Only PM10 presented percentage increase in infirmaries higher than in ICU.
Figure depicts interquartile percentage increase of the distributed lag models to infirmaries and ICU for the air pollutants included in the study. All models had the same behavior, showing high coefficients in lag 0 and 1. Only SO2 and O3 showed positive and statically significant sum of effects (lag o through lag 7).
D ISCU SSIO N
Air pollution was positively associated with increases in the daily number of MI admissions to ICU and infirmaries, and the pollutant that presented the strongest association with both outcomes was SO2.
In recent studies carried out in São Paulo, SO2 has been associated with mortality and morbidity
out-comes.12,15 Although the usual levels of SO
2 found in
São Paulo are not above the standard limits, they con-stitute a good marker of automotive emissions, the main pollution source in the city of São Paulo. The city has a population of nine million inhabitants and an automotive fleet of four million vehicles, i.e., al-most one vehicle for every two inhabitants, suggest-ing that SO2 may be a proxy for PM2,5 levels.
PM10 and O3 were also associated to the studied out-comes. The PM10 levels were associated with the daily number of MI admissions to infirmaries. This pollutant has been frequently associated to cardiovascular mor-bidity and mortality in several cities worldwide,1,11,14
and a number of hypotheses were raised about the pos-sible mechanisms that could be responpos-sible for such effects. Some of them involve inflammatory responses triggering endothelial dysfunction, atherosclerosis, and thrombosis; alterations on ion channel function in myocardial cells, and effects on the autonomic nervous system.23
The daily number of MI admissions in ICU also presented an association with ozone lev-els. This association has also been reported in the literature, although much less frequently.10
The PDL models indicated a short-time lag between the increase in air pollution and MI occurrence. The patterns of lag structure for the effects of air pollution on myocardial in-farction events have been found by other authors,2,17,24 in studies carried out in the
United States. In fact, this behavior found in observational studies is supported by experi-mental studies that have showed increases in cardiovascular events within one hour after the exposure.7
The SO2 effect on ICU admissions was higher than on infirmaries admissions. It is possible that patients admitted to ICU presented more severe disease. There are fewer vacancies than the actual demand of MI cases in the city. In the last decades, cardiovascular diseases have increased in São Paulo because of intense
Table 3 - Percentage increase and 95% CI of myocardial
infarction in hospital admissions at intensive care units and infirmaries. São Paulo, 1998-1999.
Variable Intensive Care Units Infirmary
% (95% CI) % (95% CI)
PM10 3.2 (8.6-2.2) 5.0 (9.8-0.1)
SO2 12.9 (19.4-6.4) 7.5 (13.3-1.7)
NO2 3.8 (11.4-3.8) 2.8 (9.4-3.7)
CO -0.2 (6.6-6.7) -3.1 (2.3-8.6)
O3 9.3 (17.4-1.1) 5.3 (12.7-2.1)
Figure - Percentage increases and 95% confidence intervals of myocardial
infarctions admissions to infirmaries and intensive care units due to air pollutants: 8-day distributed lag models and cumulative effects. SãoPaulo, 1998-1999.
Infirmary Intensive Care Unit
SO2 -6 -4 -2 0 2 4 6 8 10 12 14 16
Lag0 Lag1 Lag2 Lag3 Lag4 Lag5 Lag6 Lag7 SUM -5 0 5 10 15 20 25
Lag0 Lag1 Lag2Lag3 Lag4 Lag5 Lag6 Lag7SUM
O3 -4 -2 0 2 4 6 8 10 12 14
Lag0 Lag1 Lag2 Lag3 Lag4Lag5 Lag6Lag7 SUM -5 0 5 10 15 20
Lag0 Lag1 Lag2Lag3 Lag4 Lag5 Lag6 Lag7 SUM
PM10 -6 -4 -2 0 2 4 6 8 10 12
Lag0 Lag1 Lag2 Lag3 Lag4Lag5 Lag6Lag7 SUM -6 -4 -2 0 2 4 6 8 10
Lag0 Lag1 Lag2 Lag3 Lag4 Lag5 Lag6 Lag7 SUM
Infirmaries ICU
NO2 -6 -4 -2 0 2 4 6 8 10 12
Lag0 Lag1 Lag2 Lag3 Lag4 Lag5 Lag6 Lag7 SUM -6 -4 -2 0 2 4 6 8 10 12 14
urbanization process, epidemiological transition, and increased life expectance.16,* Cardiovascular diseases
has been the main causes of mortality, and MI is the third main cause in the cardiovascular group.* In this scenario MI severity has been a criterion of choice that may be determining ICU admissions. Using MI admissions to ICUs in epidemiological studies may help estimating the effect of air pollution among those patients with more severe diseases. This is a very important point that has been considered by policy decision makers. Many authors have reported
the effects of air pollutants on the cardiovascular sys-tem, but none has analyzed the relationship between MI admissions to ICU and air pollution.
Differently from respiratory diseases, the effects of air pollution on cardiovascular diseases in São Paulo are still being characterized. This study contributes to estimate the impact of air pollution in morbidity, and shows that hospital admissions to intensive criti-cal care units may be a good indicator of cardiovas-cular effects and air pollution.
*Fundação Sistema Estadual de Análise de Dados - Seade. Mortalidade por doenças do aparelho circulatório. SP Demográfico. 2003;4(2):1-4. Disponível em http://www.seade.gov.br/produtos/spdemog/PDF/fevereiro.pdf [acesso em 22 mai 2006]
REFEREN CES
1. Bateson TF, Schwartz J. Who is sensitive to the effects of particulate air pollution on mortality?: a case-crossover analysis of effect modifiers. Epidemiology. 2004;15(2):143-9.
2. Braga AL, Zanobetti A, Schwartz J. The lag structure between particulate air pollution and respiratory and cardiovascular deaths in 10 US cities. J Occup Environ Med. 2001;43(11):927-33.
3. Brunekreef B, Holgate ST. Air pollution and health. Lancet. 2002;360(9341):1233-42.
4. Devlin RB, Ghio AJ, Kehrl H, Sanders G, Cascio W. Elderly humans exposed to concentrated air pollution particles have decreased heart rate variability. Eur Respir J Suppl. 2003;40:76S-80S.
5. Dockery DW, Pope CA 3rd, Xu X, Spengler JD, Ware JH, Fay ME, et al. An association between air pollution and mortality in six US cities. N Engl J Med. 1993;329(24):1753-9.
6. Dominici F, McDermott A, Zeger SL, Samet JM. On the use of generalized additive models in time-series studies of air pollution and health. Am J Epidemiol. 2002;156(3):193-203.
7. Godleski JJ, Verrier RL, Koutrakis P, Catalano P, Coull B, Reinisch U, et al. Mechanisms of morbidity and mortality from exposure to ambient air particles. Res Rep Health Eff Inst. 2000;(91):5-103.
8. Hastie T, Tibshirani R. Generalized additive models. London: Chapman and Hall; 1990.
9. Ibald-Mulli A, Stieber J, Wichmann HE, Koenig W, Peters A. Effects of air pollution on blood pressure: a population-based approach. Am J Public Health. 2001;91(4):571-7.
10. Koken PJM, Piver WT, Ye F, Elixhauser A, Olsen LM, Portier CJ. Temperature, air pollution, and
hospitalization for cardiovascular diseases among elderly people in Denver. Environ Health Perspect. 2003;111(10):1312-7.
11. Liao D, Duan Y, Whitsel EA, Zheng ZJ, Heiss G, Chinchilli VM, et al. Association of higher levels of ambient criteria pollutants with impaired cardiac autonomic control: a population-based study. Am J Epidemiol. 2004;159(8):768-77.
12. Lin CA, Pereira LA, Nishioka DC, Conceição GM, Braga AL, Saldiva PH. Air pollution and neonatal deaths in Sao Paulo, Brazil. Braz J Med Biol Res. 2004;37(5):765-70.
13. Lin CA, Pereira LAA, Conceição GM, Kishi HS, Milani R Jr, Braga ALF, et al. Association between air pollution and ischemic cardiovascular emergency room visits. Environ Res. 2003;92(1):57-63.
14. Mann JK, Tager IB, Lurmann F, Segal M, Quesenberry CP Jr, Lugg MM, et al. Air pollution and hospital admissions for ischemic heart disease in persons with congestive heart failure or arrhythmia. Environ Health Perspect. 2002;110(12):1247-52.
15. Martins LC, Latorre M do R, Saldiva PH, Braga AL. Air pollution and emergency room visits due to chronic lower respiratory diseases in the elderly: an ecological time-series study in Sao Paulo, Brazil. J Occup Environ Med. 2002;44(7):622-7.
16. Murray CJL, Lopez AD, editors. The global burden of disease. Geneva: W orld Health Organization/ Harvard School of Public Health/The World Bank; 1996. v. 1.
17. Peters A, Dockery DW, Muller JE, Mittleman MA. Increased particulate air pollution and the triggering of myocardial infarction. Circulation.
2001;103(23):2810-5.
19. Pope CA 3rd, Verrier RL, Lovett EG, Larson AC, Raizenne ME, Kanner RE, et al. Heart rate variability associated with particulate air pollution. Am Heart J. 1999;138(5 Pt 1):890-9.
20. Santos U de P, Braga ALF, Giorgi DMA, Pereira LAA, Grupi CJ, Lin CA, et al. Effects of air pollution on blood pressure and heart rate variability: a panel study of vehicular traffic controllers in city of São Paulo, Brazil. Eur Heart J. 2005;26(2):193-200. 21. Schwartz J. Harvesting and long term exposure
effects in the relation between air pollution and mortality. Am J Epidemiol. 2000;151(5):440-8.
22. Sunyer J, Ballester F, Tertre AL, Atkinson R, Ayres JG, Forastiere F, et al. The association of daily sulfur dioxide air pollution levels with hospital admissions for cardiovascular diseases in Europe (The Aphea-II study). Eur Heart J. 2003;24(8):752-60.
23. Utell MJ, Frampton MW, Zareba W, Devlin RB, Cascio WE. Cardiovascular effects associated with air pollution: potential mechanisms and methods of testing. Inhal Toxicol. 2002;14(12):1231-47. 24. Zanobetti A, Schwartz J, Samoli E, Gryparis A,
Touloumi G, Peacock J, et al. The temporal pattern of respiratory and heart disease mortality in response to air pollution. Environ Health Perspect.
2003;111(9):1188-93.
